Dev Builds » 20181014-1819

You are viewing an old NCM Stockfish dev build test. You may find the most recent dev build tests using Stockfish 15 as the baseline here.

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 7. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host Duration Avg Base NPS Games Wins Losses Draws Elo
ncm-et-3 08:42:24 1990056 3427 1789 87 1551 +189.3 ± 8.47
ncm-et-4 08:42:19 1989560 3379 1785 90 1504 +191.6 ± 8.62
ncm-et-9 08:41:46 1934802 3228 1810 186 1232 +192.29 ± 9.7
ncm-et-10 08:42:04 1957605 3303 1772 76 1455 +197.15 ± 8.76
ncm-et-13 08:41:57 1949566 3318 1755 93 1470 +191.27 ± 8.73
ncm-et-15 08:42:09 1977361 3345 1728 89 1528 +186.24 ± 8.53
20000 10639 621 8740 +191.27 ± 3.59

Test Detail

ID Host Started (UTC) Duration Base NPS Games Wins Losses Draws Elo CLI PGN
56985 ncm-et-9 2018-10-18 16:40 00:36:31 1943454 228 129 22 77 +176.9 ± 38.72
56984 ncm-et-10 2018-10-18 16:30 00:46:42 1981068 303 156 12 135 +179.56 ± 29.08
56983 ncm-et-13 2018-10-18 16:24 00:52:18 1910551 318 172 6 140 +201.2 ± 28.25
56982 ncm-et-15 2018-10-18 16:24 00:53:05 1982322 345 187 10 148 +196.94 ± 27.71
56981 ncm-et-4 2018-10-18 16:18 00:58:49 1990191 379 203 14 162 +190.24 ± 26.59
56980 ncm-et-3 2018-10-18 16:12 01:05:06 1990350 427 213 15 199 +174.42 ± 23.77
56979 ncm-et-9 2018-10-18 15:17 01:22:14 1873355 500 295 19 186 +215.85 ± 24.99
56978 ncm-et-10 2018-10-18 15:08 01:20:41 1923117 500 260 7 233 +193.64 ± 21.63
56977 ncm-et-13 2018-10-18 15:05 01:18:17 1986721 500 259 18 223 +182.61 ± 22.55
56976 ncm-et-15 2018-10-18 15:05 01:17:36 1984835 500 258 13 229 +186.25 ± 22.07
56975 ncm-et-4 2018-10-18 14:59 01:17:32 1990033 500 264 14 222 +190.85 ± 22.52
56974 ncm-et-3 2018-10-18 14:55 01:15:49 1989558 500 244 21 235 +166.66 ± 21.92
56973 ncm-et-9 2018-10-18 13:56 01:19:36 1952064 500 310 17 173 +233.32 ± 25.95
56972 ncm-et-15 2018-10-18 13:46 01:17:36 1983262 500 259 10 231 +189.92 ± 21.86
56971 ncm-et-10 2018-10-18 13:45 01:21:47 1870858 500 270 8 222 +202.15 ± 22.33
56970 ncm-et-13 2018-10-18 13:44 01:19:35 1899718 500 276 12 212 +204.07 ± 23.08
56969 ncm-et-4 2018-10-18 13:41 01:17:33 1986091 500 256 11 233 +186.25 ± 21.78
56968 ncm-et-3 2018-10-18 13:37 01:16:40 1987666 500 276 12 212 +204.07 ± 23.08
56967 ncm-et-9 2018-10-18 12:35 01:20:20 1951916 500 271 38 191 +175.44 ± 24.7
56966 ncm-et-15 2018-10-18 12:26 01:18:55 1949622 500 252 11 237 +182.61 ± 21.54
56965 ncm-et-13 2018-10-18 12:26 01:17:27 1944651 500 265 12 223 +193.64 ± 22.4
56964 ncm-et-10 2018-10-18 12:25 01:18:48 1981227 500 258 9 233 +189.92 ± 21.71
56963 ncm-et-4 2018-10-18 12:23 01:16:52 1988770 500 273 12 215 +201.19 ± 22.89
56962 ncm-et-3 2018-10-18 12:19 01:16:43 1990348 500 268 9 223 +199.29 ± 22.3
56961 ncm-et-9 2018-10-18 11:13 01:20:15 1952218 500 276 32 192 +185.33 ± 24.64
56960 ncm-et-13 2018-10-18 11:08 01:16:27 1987360 500 259 11 230 +189.0 ± 21.95
56959 ncm-et-15 2018-10-18 11:08 01:17:45 1980285 500 258 16 226 +183.51 ± 22.33
56958 ncm-et-10 2018-10-18 11:06 01:18:11 1980132 500 283 14 203 +208.93 ± 23.72
56957 ncm-et-4 2018-10-18 11:02 01:19:16 1989244 500 260 10 230 +190.85 ± 21.92
56956 ncm-et-3 2018-10-18 11:02 01:16:08 1989243 500 270 7 223 +203.11 ± 22.23
56955 ncm-et-9 2018-10-18 09:51 01:20:35 1922936 500 278 26 196 +192.71 ± 24.36
56954 ncm-et-13 2018-10-18 09:49 01:18:25 1985472 500 272 18 210 +194.57 ± 23.35
56953 ncm-et-15 2018-10-18 09:48 01:18:36 1978880 500 249 11 240 +179.9 ± 21.37
56952 ncm-et-10 2018-10-18 09:47 01:18:22 1982948 500 266 9 225 +197.4 ± 22.18
56951 ncm-et-4 2018-10-18 09:45 01:15:54 1990507 500 263 16 221 +188.08 ± 22.63
56950 ncm-et-3 2018-10-18 09:45 01:15:59 1991931 500 256 8 236 +189.0 ± 21.5
56949 ncm-et-13 2018-10-18 08:28 01:19:28 1932492 500 252 16 232 +178.11 ± 21.98
56948 ncm-et-15 2018-10-18 08:28 01:18:36 1982321 500 265 18 217 +188.08 ± 22.92
56947 ncm-et-10 2018-10-18 08:28 01:17:33 1983890 500 279 17 204 +202.15 ± 23.72
56946 ncm-et-9 2018-10-18 08:28 01:22:15 1947676 500 251 32 217 +163.21 ± 23.1
56945 ncm-et-4 2018-10-18 08:28 01:16:23 1992090 500 266 13 221 +193.64 ± 22.55
56944 ncm-et-3 2018-10-18 08:28 01:15:59 1991297 500 262 15 223 +188.08 ± 22.48

Commit

Commit ID cb0111d3db923ec43b39f47973f7457da2b2e6c0
Author Guenther Demetz
Date 2018-10-14 18:19:46 UTC
Correctly track down pv even in fail-high case Currently we update (track up) the pv even in the fail high case. However most times in such cases the pv in the ply below remains unset because there we have value == alpha and so finally we see truncated pv's (=just one move) in fail high cases. Of course tracking down these pv's (+sending them to the gui) comes at a certian cost, but no-regression tests passed: STC: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 16300 W: 3556 L: 3424 D: 9320 http://tests.stockfishchess.org/tests/view/5b9b73500ebc592cf275ea92 LTC: LLR: 2.96 (-2.94,2.94) [-3.00,1.00] Total: 202411 W: 32734 L: 32897 D: 136780 http://tests.stockfishchess.org/tests/view/5b9baed10ebc592cf275ef6d N.B.: Digging also into qsearch was tried in another version but seemed not to pass the tests. This means that we don't always will get a pv until the very tips. No functional change
Copyright 2011–2024 Next Chess Move LLC